Interleukin 15 and granulocytes colony stimulating factor levels in relation to ICSI outcome in subfertile PCOS women


Elwiya Maternity Teaching Hospital, Baghdad, +964 770 349 0094, Iraq
Dept. of Microbiology, College of Medicine, AL-Nahrain University, Baghdad, Iraq
Medical City-Baghdad Teaching Hospital, Baghdad, Iraq

Abstract

Polycystic Ovary Syndrome constitutes a complex heterogeneous disorder, considered one of the leading cause of delayed conception in women. Serum and follicular fluid cytokines are subject of extensive studies to identify their role in subfertility with controversial results. Ovarian hormonal stimulation conducted according to GnRH antagonist protocol, then serum & follicular fluid obtained at the day of oocyte retrieval for measurement of G-CSF & IL 15 concentration measured by ELISA method, then statistical analysis is used to, retrieved oocyte number, oocyte maturity, number of 2PN oocytes, number of grade1 embryones, biochemical pregnancies and clinical pregnancy rates. Serum estradiol (E2) on hCG day was (2238.70 ±468.95 vs. 1997.40 ±291.95 pg/ml), endometrial thickness (10.80 ±1.49 vs. 10.24 ±0.98 mm), total follicle number (15.53 ±3.86 versus 12.43 ±2.25), oocyte number (13.03 ±3.10 vs. 9.83 ±1.52), mature oocyte (MII) number (9.48 ±2.45 vs. 7.75 ±1.48) mean number of fertilized oocytes (2PN) was (6.25 ±1.41 vs. 5.35 ±1.08), being higher in PCOS women compared to the control group. On day of oocytes retrieval, both median level of serum G-CSF (54.30) (35.35) and follicular G-CSF (59.00) (38.72) levels were significantly higher in PCOS than control groups, while there was no significant difference in median level of serum IL15 (59.10) (17.75) and median level of follicular IL-15(61.40) (16.00) between PCOS and control groups. In spite of higher levels of G-CSF and IL15 in serum and follicular fluid of PCOS women, a relation between G-CSF and IL15 with ovarian response could not be established in comparison to control group, however, both (G-CSF and IL15) may be a predictor of pregnancy among the same group of women both PCOS and controls.

Keywords

cytokines, IL15, G-CSF, ICSI, polycystic ovary syndrome

Introduction

Chronic anovulation is a common cause of subfertility. Majority of anovulatory women have irregular menstrual cycles and normal serum follicle-stimulating hormone (FSH) concentrations (World Health Organization WHO group 2). Polycystic ovary syndrome (PCOS) is diagnosed in approximately 60%–70% of these women, depending on the criteria used (Edmonds, 2019; Gardner, Wessman, & Colin, 2018). Accumulating evidence has demonstrated that PCOS is a low-grade pro-inflammatory disorder. An increasing number of the proinflammatory cytokines were found in the peripheral circulation and follicular fluid of retrieved oocytes obtained during IVF cycle of PCOS women (Duleba & Dokras, 2012; Qiao & Feng, 2011; Repaci, Gambineri, & Pasquali, 2011). Cytokines have been found to be directly related to maintaining the delicate balance of the hypothalamic-pituitary-ovarian axis and involved in the maintenance of normal ovarian and menstrual cycles (Field, Dasgupta, Cummings, & Orsi, 2014; Trunov, Obukhovа, Gorbenko, Shvayk, & Trunovа, 2016). Cytokines are considered the key to reproductive success, creating an immune permissive, embryo receptive environment supporting gametogenesis, fertilization, early development of the embryo, implantation of the blastocyst, and the fetus growth (Cb, A, & Ao, 2016). Their vital role in ovarian function is highly being recognized. However, the understanding of their precise function and interaction remains unclear (Cb et al., 2016). Over time, it has been shown clearly that the family of colony-stimulating factors (CSF) plays a vital role in regulating a woman's reproductive life. Furthermore, there is much evidence that CSF has a cardinal role in the early cross-talk between mother and concepts in both human and animal studies (Cavalcante, Costa, & Barini, 2015; Davari-tanha, Tehraninejad, Ghazi, & Shahraki, 2016; Eftekhar, Naghshineh, & Khani, 2018), in addition G-CSF has been found to increase in the serum up to 10 days following ovulation in successful IVF/ICSI cycles (Brenes, Sequeira, & Sosa, 2016; Rahmati et al., 2015). Interleukin-15 (IL-15) is identified as a pleiotropic cytokine that has significant roles in stimulating the production of T-helper1 cells predominant pro inflammatory cytokines, as a result promoting the proliferation of T cells and natural killer cells, in addition IL15 regulates the differentiation, development, and killing activity of the natural killer cell (Anthony & Schluns, 2015; Tamzalit. et al., 2014). So the current study was conducted to evaluate the relation of serum & follicular fluid cytokines concentration (G-CSF & IL 15) to ovarian stimulation response (Follicle number and estradiol level) and to ICSI outcome in PCOS women undergoing ICSI.

Materials and Methods

This cross-sectional study was conducted between November 2018 and April 2019 at an Infertility clinic in High Institute of Infertility Diagnosis and Assisted Reproductive Technologies at Al- Nahrain University and from Baghdad infertility & IVF private center. 80 subfertile female patients, age range 18–41 years, their BMI ranged 18.5-28.5kg ∕m2, were recruited as a study group: 40 PCOS subfertile women diagnosed as PCOS according to Rotterdam criteria (Rotterdam & ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group, 2004) and Control group: 40 non PCOS women (had other causes of subfertility). Women aged more than 42 years or with Uterine abnormality: fibroid, congenital abnormality, inadequate response of endometrial lining (endometrial thickness < 7 mm), medical & endocrine disease, e.g., advanced heart disease, thyroid disease, and hyperprolactinemia, more than 2 IVF/ICSI failure were excluded from the study. The Study approved by the local committee of the Arab Board of Medical Specialization in Iraq. Written informed consent was obtained from all patients.

Workup

Ovarian hyperstimulation was conducted according to protocol of GnRH antagonist which performed with recombinant FSH (rFSH, Gonal-F, Merck Serono, Darmstadt, Germany) or human menopausal gonadotropin (Menigon, Ferring, Saint-Prex, Switzerland), gonadotrophins was started on a menstrual day 2 or 3, gonadotrophin administration was individualized according to patient age baseline serum FSH concentration on day 3 and BMI followed by adjustment based on individual ovarian response until triggering. During ovarian stimulation, monitoring is done with serial transvaginal sonography for assessment of follicular growth, endometrial thickness, and pattern, in addition to serum estradiol (E2) level.

When the leading follicles reached a diameter of 14 mm, 0.25 mg of GnRH antagonist (cetrorelix, Cetrotide; Merck-Serono) was added daily until the day of trigger. When 2 -3 leading follicles reached a mean diameter of ≥17mm, a 250 μg of recombinant hCG (Ovitrel, Merck-Serono) was administered subcutaneously 35 hours before transvaginal oocyte retrieval. Oocytes were retrieved by transvaginal ultrasound-guided oocyte aspiration, which was done approximately 34–36 hours after hCG administration. ICSI was performed, 2–4 h later to oocyte collection. Fertilization assessment for the presence of two pronuclei (2PN) stage was conducted 16–18 h of ICSI. Embryo Transfer done under ultrasound guidance, the number of embryones transferred was limited to two or three in order to prevent multiple pregnancies. All patients received luteal support using vaginal progesterone (Crinon 8% gel, Serono, UK). Biochemical pregnancy was assessed based on the serum HCG after14 days of embryo transfer. Clinical pregnancy was assessed by the presence of intrauterine gestational sac after another 14 days after pregnancy test.

Collection and storage of the follicular fluid & serum samples

A clean and transparent (not mixed-up with blood) human FF samples were collected during oocyte retrieval. The FF samples from individual women were pooled, centrifuged at 3000 rpm for 10 minutes and collected in labelled vial then the supernatant was stored at -20°C until assayed. Venous blood samples were taken on the same day of oocyte retrieval.

Serum and follicular fluid that was obtained on the day of oocyte retrieval were estimated for cytokines levels by enzyme-linked immunosorbent assay (ELISA) technique using the diagnostic kit for IL-15 & G-CSF (Komabiotech, Seoul, Korea).

Statistical analysis

Data were collected, summarized, analyzed, and presented using statistical package for social sciences (SPSS) version 23 and Microsoft Office Excel 2010. Quantitative variables were presented as mean ± standard deviation values and compared using the Independent Samples t-test and Mann–Whitney U test. Qualitative variables were compared with Fisher’s exact or Pearson chi-square tests. Those numeric variables that are not normally distributed were expressed as median (an index of central tendency) and inter-quartile range (an index of dispersion). Pearson’s correlation coefficient (r) was applied to investigate the correlation between variables. Statistical significance was assumed with a probability error of p < 0.05.

Results and Discussion

There was a significant difference in BMI between the PCOS and control group (26.50 ±2.69 vs.25.13 ±2.16). No difference was found between the two groups regarding age and duration of infertility p-value 0.05). Regarding the duration of stimulation and total dose of gonadotrophins used, no significant difference was found between the PCOS and control group (p-value 0.05). Concerning ovarian response, there is a significant difference in oestradiol level and follicle number between both groups (p-value 0.05) (Table 2; Table 1). Also, there was a significant difference between both groups regarding endometrial thickness, Oocyte number, mature oocyte, and fertilized oocytes, being higher in PCOS group (p-value 0.05) (Table 2; Table 1). However, Grade I embryo number, biochemical and clinical pregnancy did not differ between PCOS and control group (Table 3; Table 2).

Considering cytokines levels, although there was a significant difference between both groups in serum and FF levels of G-CSF, this was not demonstrated for serum and FF levels of IL15 at p-value 0.05 (Table 4). There were increased levels of serum and FF of G-CSF and low levels of IL15 in PCOS and control women who achieved pregnancy (Table 6; Table 5). In this study, there was a significant negative correlation between serum and FF G-CSF and IL15 with BMI (r=-0.403, p=0.010 and r=-0.379, p=0.016, r=-0.217, p=0.079, r=-0.333, p=0.036, respectively) in PCOS group, but no significant correlation between cytokines levels and other clinical characteristics.

Table 1: Stimulation characteristics of PCOS and control women

Characteristic

PCOS n = 40

Control n = 40

P †

Duration of Stimulation (days)

9.48 ±0.68

9.45 ±0.71

0.873

Total Dose of gonadotrophin(IU)

1421.20 ±101.83

1417.50 ±107.15

0.873

E2 level on HCG day (pg/ml)

2238.70 ±468.95

1997.40 ±291.95

0.007 HS

Endometrial Thickness On the day of ova pick up (mm)

10.80 ±1.49

10.24 ±0.98

0.047

Table 2: Oocytes and embryo characteristics of women with PCOS and control women

Characteristic

PCOS n = 40

Control n = 40

P †

Follicle number

15.53 ±3.86

12.43 ±2.25

<0.001HS

Oocyte number

13.03 ±3.10

9.83 ±1.52

<0.001HS

Mature oocyte (MII)

9.48 ±2.45

7.75 ±1.48

<0.001HS

Fertilized oocytes (2PN)

6.25 ±1.41

5.35 ±1.08

0.002HS

Grade I embryo number

4.08 ±0.92

4.15 ±0.74

0.688

Table 3: Biochemical and clinical pregnancy outcome in PCOS and control group

Outcome

PCOS

Control

P ¥

N

%

N

%

Biochemical pregnancy

Positive

10

25.0

8

20.0

0.592

Negative

30

75.0

32

80.0

Clinical pregnancy

Positive

5

12.5

5

12.5

1.000

Negative

35

87.5

35

87.5

Table 4: Serum and follicular fluid G-CSF and IL-15 in PCOS and control group

Characteristic

PCOS n = 40

Control n = 40

P †

Serum G-CSF

54.30(35.35)

44.80(14.76)

0.0445 S

Follicular fluid G-CSF

59.00(38.72)

48.85(19.38)

0.020 S

Serum IL-15

59.10(17.75)

55.31(16.40)

0.080 NS

Follicular fluid IL-15

61.40(16.00)

58.85(19.00)

0.205 NS

Table 5: The association of serum and follicular fluid G-CSF and IL-15 levels with biochemical pregnancy outcome in PCOS and control group

Characteristic

Control

PCOS

Positive

pregnancy

n = 8

Negative

pregnancy

n = 32

Positive

pregnancy

n = 10

Negative

pregnancy

n= 30

Serum G-CSF

82.60 (17.35)

42.75 (7.20)

C

114.55 (49.88)

A

35.32 (16.63)

D

FF G-CSF

80.65 (11.65)

47.75 (6.75)

C

125.50 (58.97)

A

44.25 (17.07)

C

Serum IL-15

23.80 (37.88)

C

57.22 (21.25)

A

50.50 (9.19)

61.90 (14.32)

A

FF IL-15

32.30 (39.00)

D

60.00 (20.00)

51.54 (7.00)

C

67.45 (11.00)

A

Table 6: The association of serum and follicular fluid G-CSF and IL-15 levels with clinical pregnancy outcome in PCOS and control group

Characteristic

Control n = 40

PCOS n = 40

Positive

Pregnancy

n = 5

Negative

Pregnancy

n = 35

Positive

Pregnancy

n = 5

Negative

Pregnancy

n = 35

Serum G-CSF

88.50 (12.75)

43.33 (7.90)

C

110.20 (50.85)

A

37.61 (18.40)

C

Follicular fluid G-CSF

80.70 (18.95)

48.50 (11.00)

C

120.50 (79.55)

A

45.40 (16.00)

C

Serum IL-15

48.55 (41.61)

56.70 (21.58)

A

54.11 (7.46)

A

59.70 (19.60)

A

Follicular fluid IL-15

60.40 (43.00)

A

57.70 (19.00)

A

50.77 (9.00)

65.22 (13.00)

A

Women with PCOS are typically characterized by a large number of oocytes collected during IVF, they are frequently of poor quality, leading to decreased fertilization, cleavage and implantation rates, and poor pregnancy outcome (Gardner et al., 2018). The relationship between the immunological profile of the FF and oocyte developmental potential and implantation outcome has been uncovered (Field et al., 2014). To improve the assisted conception outcome, one essential problem that needs to be uncovered is to precisely predict oocyte and embryo developmental potential. Because the FF affects oocyte development, its composition has been studied as a potential predictor of oocyte and embryo quality (Trunov et al., 2016). In the current study PCOS group had statistically significant higher BMI than control group in spite of excluding women with BMI ≥ 28.5 kg/m2 in order to overcome the associated biochemical and endocrine abnormalities that affect the IVF outcome, this finding was similar to (Yin et al., 2015), however, other studies e.g. (Zhong, Ying, & Wu, 2012) did not show a statistical difference between these two groups. This finding can be explained by the complex interrelationships between BMI, PCOS, and metabolic abnormalities. Of particular interest are those related to the degree of Hyperinsulinemia, insulin resistance and Hyperandrogenemia which may be present in PCOS independently of BMI, all appear to be an essential pathophysiologic mechanism in the development of all metabolic complications of PCOS (Edmonds, 2019; Gardner et al., 2018).

In this study, oestradiol (E2) levels on the hCG trigger day, number of retrieved oocytes and mature oocytes, all were significantly higher in PCOS group in comparison to control group reflecting higher response to stimulation in PCOS women .Previous studies have suggested that compared to other infertility patients, PCOS patients exhibited a higher degree of ovary vascularization during the process of ovarian stimulation, which paralleled with vascular endothelial growth factor levels in the serum and follicular fluid. The ovary vascularization and VEGF levels were positively correlated with the E2 levels and numbers of oocytes achieved. In addition, VEGF can enhance the proliferation and function of granulosa cells (Gardner et al., 2018); this result was similar to previous studies. (Kahyaoglu, Yılmaz, Timur, Inal, & Erkaya, 2015; Stadtmauer & Beydoun, 2010; Yin et al., 2015). PCOS women had slightly lower fertilization rate in IVF cycles, but significantly higher fertilization rate in ICSI cycles indicating that there were certain abnormalities in zona pellucida of PCOS patients that resulted in decreased fertilization rate, ICSI could avert these abnormalities and thus increase fertilization rate (Gardner et al., 2018), in this study PCOS group had higher 2PN oocytes than the control group, this can be due to the use of ICSI in all women and the higher number of mature oocytes retrieved. This result disagrees with (Yin et al., 2015) and (Kahyaoglu et al., 2015) where fertilization rate did not show a statistically different rate between PCOS and control group and not agree with other studies (Okohue, Onuh, & Ikimalo, 2013; Rehman, Mehmood, & Ali, 2018; Zhong et al., 2012) which explain the lower fertilization rate in PCOS group by the associated endocrine abnormalities of PCOS women e.g. high LH levels and insulin resistance and internal oocytes abnormalities that can result in low rate of fertilization.

In the current study no difference is found between PCOS women and controls regarding the number of grade 1 embryo, this result is similar to (Kahyaoglu et al., 2015), however other study (Zhong et al., 2012) showed a lower number of grade 1 embryo attributed to endocrine abnormalities associated with PCOS and internal oocytes abnormalities both significantly reduce the quality of the developing embryo. The present study demonstrates similar biochemical and clinical pregnancy rates between the PCOS group and control group. In spite of high E2 levels on the hCG trigger day, number of follicles, number of retrieved oocytes and mature oocytes, fertilization (2PN oocytes) were significantly higher in the PCOS group in comparison to control group, this same result was observed in the previous studies (Okohue et al., 2013; Rehman et al., 2018; Yin et al., 2015; Zhang, Hao, & Zhuang, 2013; Zhong et al., 2012). The result of this study demonstrated that the measured FF and serum G-CSF concentrations on the day of oocyte retrieval were significantly higher in patients with PCOS undergoing IVF treatment in comparison to control group, while there was no statistical difference between these two groups regarding FF and serum IL15. Increased serum and FF levels of G-CSF in the PCOS group may favour the theory of inflammatory nature of PCOS. Chronic low-grade inflammation is currently considered to be a key feature in the pathophysiology of PCOS. In addition, increased concentrations of G-CSF in FF and serum have been associated with increased biochemical and clinical pregnancy within the same group of women undergoing ICSI (both PCOS and control). G-CSF found to play important roles in follicular maturation and ovulation exerting its biological actions through binding to specific receptors that are found on various cells including the granulosa cells of the ovary (Beydoun et al., 2009; Zhong et al., 2012). Not only G-CSF had a role in follicular growth and maturation, but also previous studies on G-CSF provided some evidence of its involvement in implantation. It seems that human decidua, to a certain degree, has a role in controlling trophoblast invasion through secretion of cytokines. G-CSF is proved to be one of those cytokines that play a vital role during the implantation window via its steady increase during early pregnancy after successful natural cycles and IVF/ICSI cycles, also it was shown convincingly that the family of CSF plays a cardinal role in the mother and conceptus early cross-talk in both animal and human studies (Cb et al., 2016; Eftekhar et al., 2018; Rahmati et al., 2015). This study was in line with accumulating literature exploring the role of serum and FF G-CSF in human reproduction. (Kahyaoglu et al., 2015), evaluate the relation of serum and FF G-CSF levels on IVF outcome in women with PCOS and women with the etiology of male factor infertility during IVF cycles. The study reveals increased levels of G-CSF both in the serum, and follicular fluid of PCOS women, a relation between G-CSF and good ovarian response or clinical pregnancy rates could not be demonstrated, and they suggest further large-scale studies to evaluate this relationship. Another study by (Niu, Ye, Xia, Feng, & Zhang, 2017), found that the low-level G-CSF levels in FF were associated with reduced top quality embryo percentage. (Gaafar et al., 2014), showed on the day of oocyte pick up, found that G-CSF was positively correlated with the number of fertilized oocytes and showed the significant positive effect to the Pregnancy outcome. However, the cytokines studied were not associated with embryo quality. In addition, (Salmassi et al., 2005), investigated G-CSF in human reproduction and found that G-CSF is involved in follicle development and may be a predictor of IVF outcome. This is not the first study on IL15 measurement in serum and FF, but it is the first to present a correlation between its serum and FF levels and clinical characteristic of women undergoing ICSI in medically stimulated cycles, including PCOS women indicated for ICSI. Increased concentrations of IL-15 in FF and serum have been associated with reduced biochemical pregnancy within the same group of women undergoing ICSI. From the foregoing, it is possible that the increased concentration of IL 15 in FF and serum is directly and/or indirectly associated with growth inhibition of the developing follicle and eventually affecting the oocyte maturity and its potential for fertilization and developing a good quality embryo. Thus, it might be related negatively with the outcome of the IVF process (Anthony & Schluns, 2015; Tamzalit. et al., 2014). Moreover, the association of increased serum and follicular IL15 levels with reduced pregnancy rates in PCOS subgroup may indicate a role for IL15 in implantation, it is found that IL15 regulation in the human endometrium is complex and IL15, as a possible stimulator for natural killer cells, found is expressed at human endometrium and first trimester decidua, and many authors recommend that a greater understanding of IL15 regulation within human endometrium is important in identifying key events of human reproduction and its role in regulating the function of uterine natural killer cells (Tamzalit. et al., 2014; Yu, Sun, & Shi, 2016). There are few researchers studied IL15 in women undergoing IVF/ICSI, all of them undergo natural IVF cycles attempts, assessing the relation between FF IL 15 concentrations and age, BMI, follicles size, retrieved oocytes maturity and IVF/ICSI cycle success and birth prediction (Lédée, Friedman, & Osipova, 2011; Spanou et al., 2018; Vujisi et al., 2006). In the current study, there was a negative correlation between BMI and serum and FF G-CSF and IL 15 in PCOS women. Other studies showed a correlation between circulating IL15 and BMI suggesting an autocrine/ paracrine nonimmune metabolic role of IL 15 in human obesity (Hingorjo, Zehra, & Saleem, 2018; Pierce, Maples, & Hickner, 2015). In spite of higher levels of G-CSF and IL15 in serum and follicular fluid of PCOS women, a relation between G-CSF and IL15 with ovarian response could not be established in comparison to the control group, however, both (G-CSF and IL15) may be a predictor of pregnancy among the same group of women both PCOS and controls.

Conclusion

The definitive knowledge of key developmental factors will provide a vision into the cellular and biochemical processes of oocyte maturation, which are important to determine its developmental potential including the unique molecular events that occur at the time of implantation and early embryonic development. The role of cytokines at different stages during assisted reproduction cycles for investigation, diagnosis, and therapy and to improve IVF success rate is still an unexplored area.